It has been discovered that antibodies and certain peptides can be used as highly specific vehicles for the delivery of drugs or radioisotopes to target organs, tumors or thrombi in vivo. Methods have been reported for the direct labeling of antibodies with radioisotopes (Huang et al., 1980, J. Nucl. Med. 21:783; Rhodes et al., in "Tumor Imaging", Burchiel and Rhodes, eds., Masson: New York, p. 111, 1982; and Sundrehagen, 1982, Eur. J. Nucl. Med. 7:549), taking advantage of partial reduction of protein disulfide linkages to generate free thiol groups capable of binding radiometals such as technetium. Because not all proteins or peptides contain readily reducible disulfides and partial reduction may alter the biological activity relative to the native molecule, it would be desirable to utilize a bifunctional radiometal chelator to form a covalent radiometal-chelator-peptide/protein conjugate capable of targeting radiometals in-vivo.
Of particular interest in nuclear medicine is the radiometal technetium-99m, which is a preferred isotope for scintigraphic imaging applications (Pinkerton et al., 1985, J. Chem. Educ. 62:965). Technetium is one of a class of metal ions that form strong coordinate bonds with sulfur-containing compounds, particularly thiols (e.g., metallothonein), but also thioureas (Koponec et al. 1977, Radiochem. Radioanal. Lett. 29:171). Koponec et al. describe the binding of technetium by thioureas to form complexes. This illustrates that thiocarbonyl functional groups bind technetium and thus may be useful as one part of a technetium binding bifunctional chelator.
The use of radiolabelled chelator conjugates may be preferable over the direct labeling systems for a number of reasons. First, the chelator may provide metal complexes of greater in vivo stability. A second advantage arises if a metabolically cleavable group is included in the linking portion of the bifunctional chelator to allow for rapid clearance and decreased accumulation of radiolabelled protein in non-targeted tissue. Lastly, direct labeling may require attachment of the radioisotope to functional groups of the protein which may lower or otherwise interfere with the in vivo targeting of the peptide or protein.
A number of bifunctional chelating agents have been reported in the scientific literature. Tolman et al. (U.S. Pat. No. 4,732,864) have described the use of the cysteine rich, metal binding protein metallothonein and metallothionein fragments conjugated to targeting molecules. However this method suffers from the fact that metallothonein is itself a large molecule and it may be difficult to purify and characterize such conjugates.
Schwartz et al. (1991, Bioconjugate Chem. 2:333) describe a series of bifunctional technetium chelators based on pyridyl hydrazines. This method of binding the radiometal is novel, however these chelators may not be useful for binding site specifically to antibodies or glycoproteins through the oxidized carbohydrate since the technetium binding hydrazide end may preferentially react with the aldehydes. Chemical solutions to this problem are not described.
Fritzberg et al. (EP 0188526; 1988, Proc. Natl. Acad. Sci. U.S.A. 85:4025) have described several examples of bifunctional dithiolate diamide technetium chelators. However, such methods for chelation of technetium are cumbersome since the compounds must be pre-chelated to technetium and then conjugated to antibodies. Also such compounds require a free thiol group for technetium chelation. Since free thiols are somewhat unstable to oxidation/dimerization and may reduce protein disulfide bonds if exposed to protein, the thiol functions in the chelator must be protected (masked) during final synthesis and then deprotected before technetium chelation.
Yokoyama et al. (U.S. Pat. No. 4,287,362; 1986, Int. J. Nucl. Med. Biol. 12:425; 1987, J. Nucl. Med. 28:1027) describe bifunctional chelators based on thiosemicarbazone derivatives of 1,2 dicarbonyl compounds. These compounds have a thiocarbonyl moiety as the technetium chelating group. Similar systems are described by Wu (EP 0306168) and are also based on thiosemicarbazone derivatives of dicarbonyl compounds. Thiosemicarbazone derivatives, which are formed by a condensation reaction through removal of water, may be susceptible to hydrolysis. Such compounds can be expected to be easily hydrolyzed in acidic aqueous media, or possibly prematurely in vivo, leading to loss of metal-chelate complex integrity. These types of hydrazone chelators are different from the present invention.
It is therefore an object of the present invention to provide a novel class of sulfur containing chelators that overcome several of these disadvantages. It is a further object to provide for specific or non-specific conjugation to a targeting molecule.
It is another object of the invention to provide sulfur-containing metal chelators that do not contain free thiol groups, and that are free of potential undesirable side reactions with proteins, especially when administered in vivo.
It is still another object of the present invention to provide sulfur-containing metal chelators that are chemically stable, and in particular that are resistant to oxidation and whose conjugates can be more resistant to hydrolysis.
Another object of the invention is to provide sulfur-containing metal chelators that contain, in some cases, in addition to a pair of thiocarbonyl chelating groups, still other pairs of chelating amino, amido, thiocarbonyl or carbonyl groups which may be useful for providing additional stability of radiometal chelates in vivo.